Search

The overall level of solar activity, and space weather response at Earth, varies within and between successive solar cycles and can be characterized by the statistics of bursts, i.e., time series excursions above a threshold. We consider nonoverlapping 1-year samples of the auroral electrojet index (AE) and the SuperMAGbased ring current index (SMR), across the last four solar cycles. These indices, respectively, characterize high latitude and equatorial geomagnetic disturbances. We suggest that average burst duration and burst return period form an activity parameter, which characterizes the fraction of time the magnetosphere spends, on average, in an active state for a given burst threshold. If the burst threshold takes a fixed value, for SMR tracks sunspot number, while for AE peaks in the solar cycle declining phase. Level crossing theory directly relates to the observed index value cumulative distribution function (cdf). For burst thresholds at fixed quantiles, we find that the probability density functions of t and R each collapse onto single empirical curves for AE at solar cycle minimum, maximum, and declining phase and for (-)SMR at solar maximum. Moreover, underlying empirical cdf tails of observed index values collapse onto common functional forms specific to each index and cycle phase when normalized to their first two moments. Together, these results offer operational support to quantifying space weather risk which requires understanding how return periods of events of a given size vary with solar cycle strength. Plain Language Summary Earth's magnetosphere and ionosphere have their own space weather. Space weather storms can cause technological problems including electrical grid damage and satellite system disruption. The overall driving of space weather follows the solar cycle of activity which has a period of approximately 11-years. Geomagnetic indices, based on magnetic field observations at the Earth's surface, provide almost continuous monitoring of magnetospheric and ionospheric activity. We analyze two geomagnetic index time series, AE and SMR, which track activity in the auroral region and around the Earth's equator, respectively. We identify bursts or excursions above thresholds in the AE and SMR time series. We find that the ratio of average burst duration to return period provides a useful activity parameter which tracks the solar cycle in a well-defined way. No two solar cycles are the same, each solar maximum has a different strength. However, the distributions of the bursts, and the observations from which they are constructed, have properties that repeat from one solar cycle to the next. These results provide constraints that could be used in model predictions for the statistics of future space weather on solar cycle scales. [ABSTRACT FROM AUTHOR]

The spatial distribution of aurora intensity is an important manifestation of solar wind‐magnetosphere‐ionosphere energy coupling process, and it oscillates with the change of space environment parameters and geomagnetic index. It is of great significance to establish an appropriate aurora intensity model for the prediction of space weather and the study of magnetosphere dynamics. Based on Ultraviolet Imager (UVI) data of Polar satellite, we constructed two auroral models by using two different neural networks, that is, the generalized regression neural network (GRNN), and the conditional generation adversarial network (CGAN). Input parameters of the models include interplanetary magnetic field, solar wind velocity and density, and the geomagnetic AE index. Output result is the spatial distribution of auroral intensity in altitude adjusted corrected geomagnetic (AACGM) coordinates. The structural similarity index (SSIM) of image quality is used as an evaluation standard of detail similarity between the prediction results of auroral intensity model and corresponding UVI images (complete similarity is 1, dissimilarity is 0, SSIM is generally considered to have good similarity if it is greater than 0.5). Based on the respective training datasets of GRNN and CGAN models, the evaluating results showed that the mean values (standard deviation) of SSIM were 0.5409 (0.0912) and 0.5876 (0.0712), respectively, so the prediction results from both models can restore the auroral intensity distribution of the original images of UVI. In addition, the value of SSIM can increase with the increase of the number of training data. Therefore, more training data will help improve the effectiveness of these models. Plain Language Summary: In the past decade, with the increase of space observation data, machine learning has been applied to space weather modeling, especially for the physical relationship is still not very clear modeling, machine learning and big data intervention can improve the applicability of the model. Although the research shows that some key parameters of space environment have a close and significant impact on aurora activity, the clear physical relationship between them is still a research problem for researchers. In this paper, the idea of directly using multiple space environment parameters to predict aurora excitation distribution is proposed, and two different aurora excitation models are established based on neural network. Moreover, these two modeling methods can be extended to other related space weather modeling. Key Points: Ultraviolet auroral intensity model driven by the interplanetary and geomagnetic parametersModeling based on different neutral network methodsStructural similarity index is used to evaluate model quality [ABSTRACT FROM AUTHOR]

The evolution of the low‐latitude magnetosphere over the substorm cycle is reconstructed based on a new high‐resolution 3D representation of the magnetic field and nearest‐neighbor data mining. The study covers radial distances 2.5–25RE and employs a record‐large pool of spacecraft data taken during 1995–2019. The magnetospheric state is quantified by four indices, representing the ground geomagnetic activity and its temporal trends in the entire ±90° range of geomagnetic latitude: the SuperMAG SMR, the midlatitude positive bay MPB, the auroral SML, and the polar cap PC index. The developed technique has been tested on specific substorm events, with the results presented in the form of 5‐min cadence diagrams and animations of the magnetic field line configurations and electric current distributions. In all the analyzed events, the initial intensification and radial expansion of the inner tail current is accompanied by a gradual stretching of the magnetic field, followed by its sudden collapse, dramatic depletion of the current beyond R∼12RE, and a large‐scale dipolarization of the field around the time of MPB peak, after which the system recovers and tends to its pre‐substorm state. Plain Language Summary: The dynamical structure of the Earth's magnetosphere during geomagnetic substorms is reconstructed, based on (a) a large multi‐year database of satellite data taken during the last quarter century, (b) a pool of concurrent ground geomagnetic activity indices, covering full range of latitudes, (c) a new magnetic field model with enhanced spatial resolution, and (d) an advanced "nearest‐neighbor" data mining approach. Based on a synthesis of the above methods and data, we explore the ability of our approach to extract maximum information from past observations and reproduce the principal phases of magnetospheric substorms in terms of time sequences of the magnetic field and electric current diagrams, from the beginning to active and recovery phase of the disturbance. Key Points: A new high‐resolution B‐field representation combined with dynamical data mining reveals magnetosphere behavior on the substorm‐time scaleFull cycle of magnetosphere evolution is reconstructed based on 25‐year archive of satellite data and a set of ground‐based activity indicesInitial growth of the distant magnetotail current and its subsequent sudden collapse during the substorm onset are consistently reproduced [ABSTRACT FROM AUTHOR]

On a geomagnetic quiet night of October 29, 2018, we captured an observational evidence of the onset of dark band structures within the field‐of‐view of an all‐sky airglow imager operating at 630.0 nm over a geomagnetic low‐mid latitude transition region, Hanle, Leh Ladakh. Simultaneous ionosonde observations over New Delhi shows the occurrence of spread‐F in the ionograms. Additionally, virtual and peak height indicate vertical upliftment in the F layer altitude and reduction in the ionospheric peak frequency were also observed when the dark band pass through the ionosonde location. All these results confirmed that the observed depletions are indeed associated with ionospheric F region plasma irregularities. The rate of total electron content index (ROTI) indicates the absence of plasma bubble activities over the equatorial/low latitude region which confirms that the observed event is a mid‐latitude plasma depletion. Our calculations reveal that the growth time of the plasma depletion is ∼2 h if one considers only the Perkins instability mechanism. This is not consistent with the present observations as the plasma depletion developed within ∼25 min. By invoking possible Es layer instabilities and associated E‐F region coupling, we show that the growth rate increases roughly by an order of magnitude. This strongly suggests that the Cosgrove and Tsunoda mechanism may be simultaneously operational in this case. Furthermore, it is also suggested that reduced F region flux‐tube integrated conductivity in the southern part of onset region created conducive background conditions for the growth of the plasma depletion on this night. Plain Language Summary: It is well known that the plasma irregularities/depletions in the ionosphere degrade the satellite‐based communication navigation signals, significantly. Thus, understanding the plausible onset condition and characteristics of these depletions are vital for the better space weather forecasting. Occurrence of plasma depletion in the equatorial and high latitude regions are associated with Generalized Rayleigh Taylor instability mechanism and their characteristics are mostly well reported. On the other hand, though the characteristics of the mid‐latitude field aligned irregularities are reported by a few investigators, there is no direct observational evidence for the onset of the mid‐latitude plasma depletion till now. Thus, the present investigation provides first optical observational evidence of in‐situ generation of plasma depletion and provides some insight on the possible background conditions which supported the onset over the geomagnetic low‐mid latitude transition region. Key Points: A remarkable observational evidence of onset of plasma depletions is captured over geomagnetic transition region in a span of 25 minIt is argued that both mid‐latitude sporadic‐E layer and Perkins instabilities are simultaneously needed to explain the growth timeIt is also proposed that reduced flux tube integrated Pedersen conductivity over south of onset location favored the growth of the bubble [ABSTRACT FROM AUTHOR]

Equatorial magnetosonic waves (EMS), together with chorus and plasmaspheric hiss, play key roles in the dynamics of energetic electron fluxes in the magnetosphere. Numerical models, developed following a first principles approach, that are used to study the evolution of high energy electron fluxes are mainly based on quasilinear diffusion. The application of such numerical codes requires statistical models for the distribution of key magnetospheric wave modes to estimate the appropriate diffusion coefficients. These waves are generally statistically modeled as a function of spatial location and geomagnetic indices (e.g., AE, Kp, or Dst). This study presents a novel dynamic spatiotemporal model for EMS wave amplitude, developed using the Nonlinear AutoRegressive Moving Average eXogenous machine learning approach. The EMS wave amplitude, measured by the Van Allen Probes, are modeled using the time lags of the solar wind and geomagnetic indices as inputs as well as the location at which the measurement is made. The resulting model performance is assessed on a separate Van Allen Probes data set, where the prediction efficiency was found to be 34.0% and the correlation coefficient was 56.9%. With more training and validation data the performance metrics could potentially be improved, however, it is also possible that the EMS wave distribution is affected by stochastic factors and the performance metrics obtained for this model are close to the potential maximum. [ABSTRACT FROM AUTHOR]

We propose a method, based on Neural Networks, that detects the nonlinear robust interplanetary solar wind variables, with varying delays, driving the coupled behavior of three geomagnetic indices (Dst, AL, and AU). As opposed to minimizing a prediction error, the method is based on degrading the prediction by distorting the inputs of the trained Neural Networks in order to highlight the most sensible drivers. We show that the z component of the magnetic field, the duskward oriented electric field, and the speed of the particles of the interplanetary medium, at particular time delays, seem to be the most efficient drivers of the three coupled geomagnetic indices. Using only the sensible or robust drivers in the model, we demonstrate that iterated predictions during geomagnetic storm are significantly improved from models that only use one of the outstanding drivers with multiple time delays. The derived robust nonlinear Neural Network model is also a significant improvement over linear approximations, specially when used as iterated predictors. Key Points: The robust interaction between solar wind and geomagnetic indices (Dst, AU, and AL) are studied using Neural Networks for hour resolutionThe robustness of the solar wind inputs that drive geomagnetic indices is evaluated byperturbing them on the trained Neural NetworksModel built with only six robust variables is 12.7% better than "bigger" models constructed with individual solar wind variables and delays [ABSTRACT FROM AUTHOR]

The present study aims to investigate the possible influence of solar/geomagnetic forcing on climate variables, such as the drought index, Danube discharge and large-scale atmospheric indices. Our analysis was performed separately for each season for two time periods, 1901–2000 and 1948–2000. The relationship between terrestrial variables and external indices was established based on the application of (1) information theory elements, namely, synergy, redundancy, total correlation, transfer entropy and (2) wavelet coherence analysis. Bandpass filtering has also been applied. The most significant signature of the solar/geomagnetic forcing in the climate variables was obtained for the data smoothed by the bandpass filter. According to our results, significant solar/geomagnetic forcing appears in the terrestrial variables with a delay of 2–3 years. [ABSTRACT FROM AUTHOR]

In situ electron density profiles obtained from Arase in the night magnetic local time (MLT) sector and from RBSP-B covering all MLTs are used to study the small-scale density irregularities present in the plasmasphere and near the plasmapause. Electron density perturbations with amplitudes >10% from background density and with time-scales less than 30-min are investigated here as the small-scale density irregularities. The statistical survey of the density irregularities is carried out using nearly 2 years of density data obtained from RBSP-B and 4 months of data from Arase satellites. The results show that density irregularities are present globally at all MLT sectors and L-shells both inside and outside the plasmapause, with a higher occurrence at L > 4. The occurrence of density irregularities is found to be higher during disturbed geomagnetic and interplanetary conditions. The case studies presented here revealed: (1) The plasmaspheric density irregularities observed during both quiet and disturbed conditions are found to coexist with the hot plasma sheet population. (2) During quiet periods, the plasma waves in the whistler-mode frequency range are found to be modulated by the small-scale density irregularities, with density depletions coinciding well with the decrease in whistler intensity. Our observations suggest that different source mechanisms are responsible for the generation of density structures at different MLTs and geomagnetic conditions. [ABSTRACT FROM AUTHOR]

Plasmaspheric hiss is one of the most important plasma waves in the Earth's magnetosphere to contribute to radiation belt dynamics by pitch-angle scattering energetic electrons via wave-particle interactions. There is growing evidence that the temporal and spatial variability of wave-particle interactions are important factors in the construction of diffusion-based models of the radiation belts. Hiss amplitudes are thought to be coherent across large distances and on long timescales inside the plasmapause, which means that hiss can act on radiation belt electrons throughout their drift trajectories for many hours. In this study, we investigate both the spatial and temporal coherence of plasmaspheric hiss between the two Van Allen Probes from November 2012 to July 2019. We find ~3,264 events where we can determine the correlation of wave amplitudes as a function of both spatial distance and time lag in order to study the spatial and temporal coherence of plasmaspheric hiss. The statistical results show that both the spatial and temporal correlation of plasmaspheric hiss decrease with increasing L-shell, and become incoherent at L > ~4.5. Inside of L = ~4.5, we find that hiss is coherent to within a spatial extent of up to ~1,500 km and a time lag up to ~10 min. We find that the spatial and temporal coherence of plasmaspheric hiss does not depend strongly on the geomagnetic index (AL*) or magnetic local time. We discuss the ramifications of our results with relevance to understanding the global characteristics of plasmaspheric hiss waves and their role in radiation belt dynamics. [ABSTRACT FROM AUTHOR]

Forecasting geomagnetic indices represents a key point to develop warning systems for the mitigation of possible effects of severe geomagnetic storms on critical ground infrastructures. Here we focus on SYM-H index, a proxy of the axially symmetric magnetic field disturbance at low and middle latitudes on the Earth's surface. To forecast SYM-H, we built two artificial neural network (ANN) models and trained both of them on two different sets of input parameters including interplanetary magnetic field components and magnitude and differing for the presence or not of previous SYM-H values. These ANN models differ in architecture being based on two conceptually different neural networks: the long short-term memory (LSTM) and the convolutional neural network (CNN). Both networks are trained, validated, and tested on a total of 42 geomagnetic storms among the most intense that occurred between 1998 and 2018. Performance comparison of the two ANN models shows that (1) both are able to well forecast SYM-H index 1 h in advance, with an accuracy of more than 95% in terms of the coefficient of determination R²; (2) the model based on LSTM is slightly more accurate than that based on CNN when including SYM-H index at previous steps among the inputs; and (3) the model based on CNN has interesting potentialities being more accurate than that based on LSTM when not including SYM-H index among the inputs. Predictions made including SYM-H index among the inputs provide a root mean squared error on average 42% lower than that of predictions made without SYM-H. [ABSTRACT FROM AUTHOR]

In this study, ion data from the Helium, Oxygen, Proton, and Electron (HOPE) spectrometers onboard Van Allen Probes reveal the existence of wedge‐like structures of O+, He+, and H+ ions deep in the inner magnetosphere. The behaviors of the wedge‐like structures in terms of temporal evolution, spatial distribution, upper energy limit, as well as dependence on solar wind and different geomagnetic indices are investigated from both event studies of several consecutive orbits on 3 February 2013 and the subsequent statistical analyses using 4 years of data. Unlike the dominant distribution at L=4–8 in the dayside observed by the polar orbit satellites in previous studies, the wedge‐like structures deep in the equatorial plane of the inner magnetosphere are found mostly at the Mcllwain L shells of L=2–5 and have a preferential location in the duskside and nightside. The O+ and He+ structures can extend to smaller L shells with higher upper energy limits than the H+ structures, while the upper energy limits of all these particle species show a similar variation tendency with respect to magnetic local time (MLT) and L. Observations indicate that these wedge‐like structures are probably attributed to fresh substorm injections from the outer region. Key Points: Van Allen Probe observations reveal the existence of wedge‐like ion structures mainly at the L shells of L=2–5Structures have a preferential location in the duskside and nightside, and the O+ and He+ structures have higher upper energy limits than H+The formation of these structures are probably attributed to substorm injections drifting earthward from the outer region [ABSTRACT FROM AUTHOR]

Increased ionization in the auroral oval leads to the absorption of high‐frequency radio waves in the auroral zone, or auroral absorption. Auroral absorption is typically characterized by global geomagnetic activity indices, such as the Kp index. In this paper the hourly range of the magnetic field (HR) is examined as an alternative to the 3‐hr Kp index for describing the dynamic and localized features of auroral absorption represented by the hourly range of absorption (HRA). Kp, magnetometer, and riometer data were examined for a 3‐year period for stations spread across typical auroral latitudes. A general linear relationship was shown to exist between Kp and LOG10(HRA) for Kp < 4; for Kp ≥ 4 the correlation was weaker. A stronger linear correlation was demonstrated between LOG10(HRA) and LOG10(HR) for HR > 50 nT, characterized by a correlation coefficient of R = 0.63. Increased variability in the relationship between HRA and Kp was attributed to the following factors: the variability of the magnetic field within the 3‐hr window characterized by the Kp index, which was better represented by a 1‐hr HR; the dependence of the Kp index on subauroral magnetic data, which is not subject to the geomagnetic variations typically experienced within the auroral region; and reduced statistics for Kp > 6. Key Point: Hourly range of the magnetic field is stronger than Kp for characterizing auroral absorption [ABSTRACT FROM AUTHOR]

The flux of > 2 MeV electrons at geosynchronous orbit is used by space weather forecasters as a key indicator of enhanced risk of damage to spacecraft in low, medium, or geosynchronous Earth orbits. We present a methodology that uses the amount of time a single input data set (solar wind data or geomagnetic indices) exceeds a given threshold to produce deterministic and probabilistic forecasts of the >2 MeV flux at GEO exceeding 1,000 or 10,000 cm−2 s−1 sr−1 within up to 10 days. By comparing our forecasts with measured fluxes from GOES 15 between 2014 and 2016, we determine the optimum forecast thresholds for deterministic and probabilistic forecasts by maximizing the receiver‐operating characteristic (ROC) and Brier skill scores, respectively. The training data set gives peak ROC scores of 0.71 to 0.87 and peak Brier skill scores of −0.03 to 0.32. Forecasts from AL give the highest skill scores for forecasts of up to 6 days. AL, solar wind pressure, or SYM‐H give the highest skill scores over 7–10 days. Hit rates range over 56–89% with false alarm rates of 11–53%. Applied to 2012, 2013, and 2017, our best forecasts have hit rates of 56–83% and false alarm rates of 10–20%. Further tuning of the forecasts may improve these. Our hit rates are comparable to those from operational fluence forecasts, that incorporate fluence measurements, but our false alarm rates are higher. This proof‐of‐concept shows that the geosynchronous electron flux can be forecast with a degree of success without incorporating a persistence element into the forecasts. Plain Language Summary: Spacecraft that orbit the Earth 36,000 km above the equator take 24 hr to orbit the Earth, meaning they stay above the same point on the Earth's all the time. These "geosynchronous" orbits are incredibly useful, enabling satellites to have constant contact with the ground. As of 31 March 2019, over 500 spacecraft are in geosynchronous orbit (https://www.ucsusa.org). Geosynchronous orbit is also on the edge of one of the most hazardous regions of space around the Earth—the Van Allen Radiation Belts. These belts contain highly energetic particles capable of damaging spacecraft in geosynchronous orbit and are highly variable. As such, predicting when the radiation belts at geosynchronous orbit are dangerous is a key concern in space weather. In this study, we create forecasts of the radiation belts based on simple measurements of upstream and local conditions of Earth's space environment. These simple forecasts provide the probabilities of high‐risk events at geosynchronous orbit over periods of up to 10 days, providing an interesting new mechanism for forecasting the conditions in near‐Earth space as well as helping us to understand the factors that control the dynamics of the radiation belts. Key Points: A simple method for deterministic or probabilistic forecasting of the geosynchronous >2 MeV electron flux exceeding set levels is examinedForecasts are based on the amount of time a single input variable exceeds a set threshold and do not include measured geosynchronous fluxAL, SYM‐H, and solar wind pressure provide the most skilful forecasts when analyzed using standard forecasting metrics [ABSTRACT FROM AUTHOR]

One of the main goals when studying Space Weather is to characterize extreme events occurrences and related characteristics. To do so, dedicated statistical methods from the so‐called extreme value analysis (EVA) field have been developed. In this study we used Ca index, derived from aa, in order to characterize geoeffectiveness from the radiation belts point of view with a 150‐year‐long data set. The analysis performed in this study thus focuses on this newsworthy index to provide clues on the reliability of EVA methods. The first main result we present here is that the 1‐in‐10‐, 1‐in‐50‐, and 1‐in‐100‐year events, respectively, match Ca values of 100.39, 131.39, and 142.84 nT. Consequently, the only 1‐in‐100 event observed during the Space Era would be the "Halloween Storm" in 2003 that reached a Ca value of 147.6 nT. The second main result highlighted in this work is that performing the same analysis with shorter subsets (20 years long) can give significantly different results for two reasons. The first reason is that some short time periods do not display the same distribution of events as the full period. The second reason is that the choice of the correct threshold (when using a Peaks Over Threshold approach) is made difficult with a short data set and leads to inaccurate results. This is a strong result as for accurate estimation of the induced effects of extreme events in radiation belts, we may only rely on short flux data sets from one or another mission (mostly shorter than 20 years). Key Points: Ca is a new time‐integrated geomagnetic index based on aa, defined to characterize geoeffectiveness from the radiation belts point of viewDedicated extreme value analysis methods are applied to a 150‐year‐long data set of geomagnetic events based on Ca indexThe use of the same method on data sets based on periods covered by just one space mission can give different results for two main reasons [ABSTRACT FROM AUTHOR]

For decades, geomagnetic indices have been used extensively to parameterize space weather events, as input to various models and as space weather specifications. The auroral electrojet (AE) index and disturbance storm time index (DST) are two such indices that span multiple solar cycles and have been widely studied. The production of improved spatial coverage analogs to AE and DST is now possible using the SuperMAG collaboration of ground‐based magnetometers. SME is an electrojet index that shares methodology with AE. SMR is a ring current index that shares methodology with DST. As the number of magnetometer stations in the SuperMAG network increases over time, so does the spatial resolution of SME and SMR. Our statistical comparison between the established indices and their new SuperMAG counterparts finds that, for large excursions in geomagnetic activity, AE systematically underestimates SME for later cycles. The difference between distributions of recorded AE and SME values for a single solar maximum can be of the same order as changes in activity seen from one solar cycle to the next. We demonstrate that DST and SMR track each other but are subject to an approximate linear shift as a result of the procedure used to map stations to the magnetic equator. We explain the observed differences between AE and SME with the assistance of a simple model, based on the construction methodology of the electrojet indices. We show that in the case of AE and SME, it is not possible to simply translate between the two indices. Plain Language Summary: Space weather events can cause disturbances in the Earth's magnetosphere and ionosphere. Severe disturbances can cause disruption to electrical power systems, aviation, communication systems, and satellite systems. Magnetometer stations on the ground are used to monitor and specify changes in the magnetosphere‐ionosphere system. Geomagnetic indices based on measurements from these stations are used extensively, and they have been recorded for many decades. Two examples are AE and DST, which are indices designed to measure the evolution and intensity of the auroral electrojets and the ring current, respectively. The SuperMAG collaboration has made new versions of these indices available. They are based on a larger number of magnetometer stations than the original AE and DST indices. We carry out a statistical comparison between the traditional and updated indices to identify how improved spatial resolution affects the indices. Key Points: We present a statistical comparison of AE and DST with SME and SMR, their higher spatial resolution SuperMAG counterpartsAE systematically undersamples when compared to SME for later solar cycles; AE and SME differ at same scale as cycle‐to‐cycle variabilityDST and SMR track each other with a small systematic linear shift [ABSTRACT FROM AUTHOR]

Forecasting relativistic electron fluxes at geostationary Earth orbit (GEO) has been a long‐term goal of the scientific community, and significant advances have been made in the past, but the relation to the interior of the radiation belts, that is, to lower L‐shells, is still not clear. In this work we have identified 60 relativistic electron enhancement events at GEO to study the radial response of outer belt fluxes and the correlation between the fluxes at GEO and those at lower L‐shells. The enhancement events occurred between 1 October 2012 and 31 December 2017 and were identified using Geostationary Operational Environmental Satellite (GOES) 15 >2 MeV fluxes at GEO, which we have used to characterize the radial response of the radiation belt, by comparing to fluxes measured by the Van Allen probes Energetic Particle, Composition and Thermal Plasma Suite Relativistic Electron‐Proton Telescope (ECT‐REPT) between 2.55.0 and generally similar for L>4.5. Post‐enhancement maximum fluxes show a remarkable correlation for all L>4.0 although the magnitude of the pre‐existing fluxes on the outer belt plays a significant role and makes the ratio of pre‐enhancement to post‐enhancement fluxes less predictable in the region 4.0

The Van Allen radiation belts in the magnetosphere have been extensively studied using models based on radial diffusion theory, which is derived from a quasi‐linear approach with prescribed inner and outer boundary conditions. The 1D diffusion model requires the knowledge of a diffusion coefficient and an electron loss timescale, which is typically parameterized in terms of various quantities such as the spatial (L) coordinate or a geomagnetic index (e.g., Kp). These terms are typically empirically derived, not directly measurable, and hence are not known precisely, due to the inherent nonlinearity of the process and the variable boundary conditions. In this work, we demonstrate a probabilistic approach by inferring the values of the diffusion and loss term parameters, along with their uncertainty, in a Bayesian framework, where identification is obtained using the Van Allen Probe measurements. Our results show that the probabilistic approach statistically improves the performance of the model, compared to the empirical parameterization employed in the literature. Key Points: We present the first application of Bayesian parameter estimation to the problem of quasi‐linear radial diffusion in the radiation beltThe Bayesian approach allows the problem to be cast in probabilistic terms and for ensemble simulations to be runAn improved accuracy is demonstrated when compared against standard deterministic models [ABSTRACT FROM AUTHOR]

We apply singular spectrum analysis in order to identify trends and quasi‐periodic oscillations in the aa and Dst series of geomagnetic activity. We also analyze the sunspot number International SunSpot Number (ISSN) and the number of polar faculae Polar Faculae (PF). Singular spectrum analysis provides the eigenvalues and therefore trends and oscillatory components of the four series. ISSN is dominated by a trend (the Gleissberg cycle), followed by 10.6, 35.5 years, two ~8‐year components, 21.4 and 5.3 year. aa shows the same trend, a ~47‐year component, then 10.8, 32.3, 21.8, and a series of three close components at 10.6, 12.2, and 9.2 years, followed by a 6 month seasonal component. PF is dominated by the 20.7‐year period, followed by 10.2, 8.3, 41, and 31 years, then a 5.2 year component. Dst is dominated by a trend, then a strong 6‐month component; next are found a 47‐year component, the 10.6 years and a second seasonal line at 1 year. The ~22‐, ~11‐, and ~5.5‐year components are common to the four indices. These "pseudo harmonic" components are evidence of solar activity. Singular spectrum analysis identifies components that vary in frequency and amplitude. The phase relationships of any two components over time can be studied in detail. An illustration is given by the remarkable phase coherency of the 5.3‐year component. But the components are neither truly periodical nor exact multiples of each other. These differences reflect the complex mechanisms that govern solar‐terrestrial relationships. Key Points: Singular spectrum analysis reveals trends and quasi‐periodic oscillations in the aa and Dst series of geomagnetic activityThe ~22‐, ~11‐, and ~5.5‐year components are common to the four indices. These "pseudo harmonic" components are evidence of solar activitySSA components vary in frequency and amplitude. The phase relationships of any two components over time can be studied in detail [ABSTRACT FROM AUTHOR]

Current algorithms for the real‐time prediction of the Kp index use a combination of models empirically driven by solar wind measurements at the L1 Lagrange point and historical values of the index. In this study, we explore the limitations of this approach, examining the forecast for short and long lead times using measurements at L1 and Kp time series as input to artificial neural networks. We explore the relative efficiency of the solar wind‐based predictions, predictions based on recurrence, and predictions based on persistence. Our modeling results show that for short‐term forecasts of approximately half a day, the addition of the historical values of Kp to the measured solar wind values provides a barely noticeable improvement. For a longer‐term forecast of more than 2 days, predictions can be made using recurrence only, while solar wind measurements provide very little improvement for a forecast with long horizon times. We also examine predictions for disturbed and quiet geomagnetic activity conditions. Our results show that the paucity of historical measurements of the solar wind for high Kp results in a lower accuracy of predictions during disturbed conditions. Rebalancing of input data can help tailor the predictions for more disturbed conditions. Key Points: Short‐term predictions should be made with solar wind data and long‐term predictions with recurrence forecastThe accuracy of predictions is much lower during disturbed geomagnetic conditionsRebalancing allows us to improve the accuracy of predictions during disturbed geomagnetic conditions [ABSTRACT FROM AUTHOR]

Eruptive events on the Sun have an impact on the immediate surroundings of the Earth. Through induction of electric currents, they also affect Earth‐bound structures such as the electric power transmission networks. Inspired by recent studies we investigate the correlation between the disturbances recorded in 12 years in the maintenance logs of the Czech electric power distributors with the geomagnetic activity represented by the K index. We find that in case of the data sets recording the disturbances on power lines at the high and very high voltage levels and disturbances on electrical substations, there is a statistically significant increase of anomaly rates in the periods of tens of days around maxima of geomagnetic activity compared to the adjacent minima of activity. There are hints that the disturbances are more pronounced shortly after the maxima than shortly before the maxima of activity. Our results provide indirect evidence that the geomagnetically induced currents may affect the occurrence rate of anomalies registered on power‐grid equipment even in the midlatitude country in the middle of Europe. A follow‐up study that includes the modeling of geomagnetically induced currents is needed to confirm our findings. Key Points: We compare the series of disturbances recorded in the Czech power grid with the geomagnetic activityThe comparison is done in a statistical sense by considering only possible time scale of tens of daysWe find indications that the midlatitude power grid may be also affected by space weather events [ABSTRACT FROM AUTHOR]

The numerous recent breakthroughs in machine learning make imperative to carefully ponder how the scientific community can benefit from a technology that, although not necessarily new, is today living its golden age. This Grand Challenge review paper is focused on the present and future role of machine learning in Space Weather. The purpose is twofold. On one hand, we will discuss previous works that use machine learning for Space Weather forecasting, focusing in particular on the few areas that have seen most activity: the forecasting of geomagnetic indices, of relativistic electrons at geosynchronous orbits, of solar flares occurrence, of coronal mass ejection propagation time, and of solar wind speed. On the other hand, this paper serves as a gentle introduction to the field of machine learning tailored to the Space Weather community and as a pointer to a number of open challenges that we believe the community should undertake in the next decade. The recurring themes throughout the review are the need to shift our forecasting paradigm to a probabilistic approach focused on the reliable assessment of uncertainties, and the combination of physics‐based and machine learning approaches, known as gray box. Plain Language Summary: In the last decade, machine learning has achieved unforeseen results in industrial applications. In particular, the combination of massive data sets and computing with specialized processors (graphics processing units, or GPUs) can perform as well or better than humans in tasks like image classification and game playing. Space weather is a discipline that lives between academia and industry, given the relevant physical effects on satellites and power grids in a variety of applications, and the field therefore stands to benefit from the advances made in industrial applications. Today, machine learning poses both a challenge and an opportunity for the space weather community. The challenge is that the current data science revolution has not been fully embraced, possibly because space physicists remain skeptical of the gains achievable with machine learning. If the community can master the relevant technical skills, they should be able to appreciate what is possible within a few years time and what is possible within a decade. The clearest opportunity lies in creating space weather forecasting models that can respond in real time and that are built on both physics predictions and on observed data. Key Points: Machine learning (ML) has enabled advances in industrial applications; space weather researchers are adopting and adapting ML techniquesThis introduction to machine learning concepts is tailored for the Space Weather community, but applicable to many other communitiesThis introduction describes forecasting opportunities in a gray‐box paradigm that combines physics‐based and machine learning approaches [ABSTRACT FROM AUTHOR]

This study addresses the question whether to what extent (if any) light-trap catch of the harmful pest, Helicoverpa armigera (Hübner, 1805) (Lepidoptera: Noctuidae) depends on the Moon phases and the geomagnetic horizontal component (H-index). Therefore, daily relative catch data were assigned to the daily values of geomagnetic field above 0.2125−10 T (H-index). We correlated the daily catch results to the daily values of geometric H-index. The numbers of specimens caught and classified according to generation were calculated relative to catch values. Relative catch data were divided according to the Moon phase angle around the four Moon quarters. The relative daily catch data were assigned to the daily values of geomagnetic H-index. We correlated the daily catch results to the daily values of geometric H-index in four moon phases. More abundant catch corresponded to higher H-index values in the New Moon period. Approaching Full Moon led to increasing catch correlating to low H-index value, but there decreasing catch coincided with higher H-index values. There was also decreasing catch at the increasing H-index values in the First Quarter and Last Quarter period. The light trap catch of the H. armigera was seemingly influenced by both the moon phases and the horizontal component of geomagnetism. [ABSTRACT FROM AUTHOR]

A frequency‐tunable resonance scattering lidar with high temporal/vertical resolutions (1 min/15 m) observed sporadic calcium ion (Ca+s) layers at ~100 km over Tachikawa (geographical/geomagnetic latitude: 35.7°N/27.1°N), Japan, on 21–22 August 2014. Simultaneously, sporadic E (Es) parameters and medium‐scale traveling ionospheric disturbances (MSTIDs) were observed by an ionosonde and Global Navigation Satellite System receiver network, GEONET, respectively. The maximum densities of the Ca+ and electrons in the Es layer had a strong positive correlation. As observation started ~23:30 LT, the Ca+s layer and the associated Es layer descended at ~2.8 km/hr with density irregularities including Kelvin‐Helmholtz billow‐like structures suggesting the presence of background neutral wind shear and instability. And the total electron content variations showed large amplitude associated with the MSTIDs at an altitude of 300 km in synchronization with the Ca+ column abundance surges at 100 km over Tachikawa; in their respective E and F region locations connected by geomagnetic field line these irregularities are found to vary in phase. At ~02:00 LT, the Ca+s layer stopped descending at ~100 km due to larger ion‐neutral collision frequency in the lower altitudes and resided there quietly until sunrise; both Ca+ column abundance enhancements and the large total electron content variation disappeared as the descent of the Ca+s layer stopped, implicating that the MSTID structure cannot be sustained without the density irregularities of the Es layer. This is the first synchronous observation of the coupling between the Es density irregularities and the MSTIDs in the F region along a common magnetic flux tube. Plain Language Summary: The irregular structures of the sporadic E (Es) layer appearing in altitudes of 90–130 km were observed by a calcium ion (Ca+) resonance scattering lidar and an ionosonde. Simultaneously, medium‐scale traveling ionospheric disturbances (MSTIDs) in the F region (150‐ to 500‐km altitude) were observed by the Global Navigation Satellite System receiver array in Japan. Though the electron column abundance variation in the Es irregularities is more than an order of magnitude smaller than that in the MSTIDs, we observed that the MSTIDs decayed with the disappearance of the irregular structure in the Es layer. This observation clearly revealed for the first time that the plasma density irregularities in the E region were coupled to those in the F region by the geomagnetic field line as theory predicted. Key Points: The irregularities of the Es layer and the MSTIDs were observed simultaneously by a Ca+ lidar, an ionosonde, and a GNSS receiver networkThe simultaneous observation clearly revealed how the Es layer and the F region ionosphere are coupledThis is the first synchronous observation of the coupling between the density irregularities of the Es layer and the MSTIDs in the F region [ABSTRACT FROM AUTHOR]

Kinetic Alfvén wave (KAW) and whistler wave play a key role in the framework of turbulence and reconnection. There are lot of observations of these wave in the magnetopause region. The present paper deals with the nonlinear evolution of KAW and a weak whistler wave through the preexisting magnetic reconnection site. For this study the dynamical evolution equations are derived by taking into account the ponderomotive force driven density modification and magnetic field fluctuations due to shear field modeled by the Harris sheet. Furthermore, these equations have been solved numerically as well as semianalytically. For numerical integrations we have used the pseudospectral method and finite difference method and for semianalytically Runge Kutta method. Simulated results have shown the evolution of coherent structures or current sheets, which are capable to energy transfer efficiently. These structures have scale size around ion gyroradius as well as electron inertial length as calculated analytically. At a later time the chaotic structures arise in this reconnection site. This gives the signature of turbulence generation. Therefore, the magnetic power spectrum with scaling is also presented in this manuscript and their relevance with the observed spectrum (calculated from the Cluster data (Chaston, 2008, https://doi.org/10.1029/2008GL033601)) is also pointed out. Key Points: Role of KAW and Whistler waves in the reconnection site is very importantA model is developed to study the effects of background Harris field and KAW nonlinearity in evolution of KAW and weak whistler wave in magnetic reconnection siteLocalized structures have scale sizes varying from ion gyroradious to electron skin depth [ABSTRACT FROM AUTHOR]

Subauroral polarization streams (SAPS) prefer geomagnetically disturbed conditions and strongly correlate with geomagnetic indexes. However, the temporal evolution of SAPS and its relationship with dynamic and structured ring current and particle injection are still not well understood. In this study, we performed detailed analysis of temporal evolution of SAPS during a moderate storm on 18 May 2013 using conjugate observations of SAPS from the Van Allen Probes (VAP) and the Super Dual Auroral Radar Network (SuperDARN). The large‐scale SAPS (LS‐SAPS) formed during the main phase of this storm and decayed due to the northward turning of the interplanetary magnetic field. A mesoscale (approximately several hundreds of kilometers zonally) enhancement of SAPS was observed by SuperDARN at 0456 UT. In the conjugate magnetosphere, a large SAPS electric field (∼8 mV/m) pointing radially outward, a local magnetic field dip, and a dispersionless ion injection were observed simultaneously by VAP‐A at L shell = 3.5 and MLT = 20. The particle injection observed by VAP‐A is likely associated with the particle injection observed by the Geostationary Operational Environmental Satellite 15 near 20 MLT. Magnetic perturbations observed by the ground magnetometers and flow reversals observed by SuperDARN reveal that this mesoscale enhancement of SAPS developed near the Harang reversal and before the substorm onset. The observed complex signatures in both space and ground can be explained by a two‐loop current wedge generated by the perturbed plasma pressure gradient and the diamagnetic effect of the structured ring current following particle injection. Key Points: Mesoscale enhancement of SAPS was observed by VAP and SuperDARN on top of the existing large‐scale SAPSMesoscale enhancement is associated with energetic ion flux increase, energetic electron flux decrease, and local magnetic field dipMesoscale enhancement of SAPS and equatorward flow burst developed near the Harang reversal [ABSTRACT FROM AUTHOR]

Magnetospheric substorms represent key explosive processes in the interaction of the Earth's magnetosphere with the solar wind, and their understanding and modeling are critical for space weather forecasting. During substorms, the magnetic field on the nightside is first stretched in the antisunward direction and then it rapidly contracts earthward bringing hot plasmas from the distant space regions into the inner magnetosphere, where they contribute to geomagnetic storms and Joule dissipation in the polar ionosphere, causing impressive splashes of aurora. Here we show for the first time that mining millions of spaceborne magnetometer data records from multiple missions allows one to reconstruct the global 3‐D picture of these stretching and dipolarization processes. Stretching results in the formation of a thin (less than the Earth's radius) and strong current sheet, which is diverted into the ionosphere during dipolarization. In the meantime, the dipolarization signal propagates further into the inner magnetosphere resulting in the accumulation of a longer lived current there, giving rise to a protogeomagnetic storm. The global 3‐D structure of the corresponding substorm currents including the substorm current wedge is reconstructed from data. Plain Language Summary: Using several millions of historical magnetometer records and data mining techniques, we form virtual spacecraft constellations of tens of thousands of spacecraft to reconstruct the global shape of the terrestrial magnetosphere at the moments of its most dramatic reconfigurations responsible for major space weather disturbances. Key Points: Substorm tail current sheet thinning and dipolarization are reproduced using novel data mining techniqueGlobal 3‐D structure of substorm currents including the substorm current wedge is reconstructed from dataSubstorms contribute to an accumulation of a longer‐lived thick current in the innermost part of the magnetosphere [ABSTRACT FROM AUTHOR]

Statistical models have been developed for predicting the behavior of the coupled high‐latitude ionosphere‐thermosphere system. The modeled parameters were the F‐layer peak electron density, plasma structuring, ion temperature, neutral temperature, and the difference between these temperatures, which is a key term in the Joule heating equation. Ionospheric measurements from the European Incoherent Scatter Svalbard Radar and neutral atmosphere measurements from the colocated University College London Fabry‐Perot Interferometers have been made across a solar cycle. These data were all acquired during nighttime conditions as the observations with the Fabry‐Perot Interferometers are restricted to such times. Various geophysical proxies were tested to represent the processes that influence the modeled parameters. The dominant geophysical proxy for each modeled parameter was then determined. Multivariate models were also developed showing the combinations of parameters that best explained the observed variability. A comparison with climatology showed that the models give an improvement in every case with skill scores based on the mean square error of up to 0.88. Plain Language Summary: The upper atmosphere of the Earth is a mixture of partially ionized plasma (the ionosphere) and neutral gases (the thermosphere). The plasma is comprised of charged particles which are subject to electromagnetic forces, whereas the neutral thermosphere is not. These two populations are the coupled ionosphere‐thermosphere. Plasma density in the ionosphere is controlled by plasma production mechanisms such as solar ultraviolet illumination, plasma loss mechanisms such as collision induced recombination, and transport mechanisms such as variability in the geomagnetic field. Of particular interest is the process of heat transfer from ionospheric plasma to the neutral thermosphere, so‐called Joule heating. Combined with geophysical information, such as the Kp index, variability in the interplanetary magnetic field, and season, this study presents a series of statistical linear models that rank various physical effects in terms of how strongly they influence the behavior of the coupled ionosphere‐thermosphere at high latitude during polar darkness. These models also show which combinations of physical effects enable predictions to be made of observed ionosphere‐thermosphere behavior, to a greater accuracy than by a climatological approach. This study was made possible using ionosphere data collected from the European Incoherent Scatter Svalbard radar with neutral thermosphere observations from a colocated Fabry‐Perot Interferometer. Key Points: Linear modeling of variability in the high‐latitude coupled ionosphere‐thermosphere is a more accurate predictor than climatologyUnivariate linear models demonstrate the relative influences of different geophysical proxies on ionosphere‐thermosphere variability [ABSTRACT FROM AUTHOR]

Models that predict the Kp and Dst indices are evaluated using solar wind data at L1. The models consist of ensembles of neural networks that have been developed using ACE Level 2 data from the period 1998–2015. The use of ensembles is motivated by the difficulty of generating functions that generalize well in regions of the input‐output space that are poorly sampled, which typically occurs during stronger events. Since the launch of the DSCOVR spacecraft, providing measurements from about August 2016, new and independent data have become available to test the models. ACE Level 2 data for almost the same period are also available, representing another independent set collected after 2015. We evaluate the models using plots and various statistical measures. We also study the performance of the predictions for lead times up to 3 hr. The results show that the models perform better when using ACE or cleaned DSCOVR data compared to the real‐time DSCOVR data and that lead times of L1‐Earth travel time plus a maximum of 1 hr are possible. As the models use only solar wind for their inputs, and the temporal dynamics of Kp and Dst are very different, we see significant differences in the error distributions that we believe are related to long‐term changes in Dst that are not captured by the Dst prediction model. Key Point: Solar wind‐driven Kp and Dst index models were evaluated on ACE and DSCOVR data [ABSTRACT FROM AUTHOR]

Stream interaction regions (SIRs) are important sources of geomagnetic storms. In this work, we first extend the end time of the widely used SIR catalog developed by Jian et al. (2006, https://doi.org/10.1007/s11207-006-0132-3), which covered the period from 1995 to 2009, to the end of 2016. Based on this extended SIR catalog, the geoeffectiveness of SIRs is discussed in detail. It was found that 52% of the SIRs caused geomagnetic storms with Dstmin ≤−30 nT, but only 3% of them caused intense geomagnetic storms with Dstmin ≤−100 nT. Furthermore, we found that 10 of the intense geomagnetic storms caused by SIRs were associated with complex structures due to interactions between SIRs and interplanetary coronal mass ejections (ICMEs). In such a structure, an ICME is embedded in the SIR and located between the slow and fast solar wind streams. In addition, we found that the geoeffectiveness of SIRs interacting with ICMEs is enhanced. The possibility of SIR‐ICME interaction structures causing geomagnetic storms is markedly higher than that of isolated SIRs or isolated ICMEs. In particular, the geoeffectiveness of SIR‐ICME interaction structures is similar to that of the Shock‐ICME interaction structures, which have been demonstrated to be the main causes of geomagnetic storms. Key Points: The near‐Earth SIR catalog developed by Jian et al. (2006) has been extended to the end of 2016Fifty‐two percent of the SIRs caused geomagnetic storms, but only 3% of the SIRs caused intense geomagnetic stormsSIRs that interact with ICMEs exhibit significantly enhanced geoeffectiveness [ABSTRACT FROM AUTHOR]

Geomagnetic indices are convenient quantities that distill the complicated physics of some region or aspect of near‐Earth space into a single parameter. Most of the best‐known indices are calculated from ground‐based magnetometer data sets, such as Dst, SYM‐H, Kp, AE, AL, and PC. Many models have been created that predict the values of these indices, often using solar wind measurements upstream from Earth as the input variables to the calculation. This document reviews the current state of models that predict geomagnetic indices and the methods used to assess their ability to reproduce the target index time series. These existing methods are synthesized into a baseline collection of metrics for benchmarking a new or updated geomagnetic index prediction model. These methods fall into two categories: (1) fit performance metrics such as root‐mean‐square error and mean absolute error that are applied to a time series comparison of model output and observations and (2) event detection performance metrics such as Heidke Skill Score and probability of detection that are derived from a contingency table that compares model and observation values exceeding (or not) a threshold value. A few examples of codes being used with this set of metrics are presented, and other aspects of metrics assessment best practices, limitations, and uncertainties are discussed, including several caveats to consider when using geomagnetic indices. Plain Language Summary: One aspect of space weather is a magnetic signature across the surface of the Earth. The creation of this signal involves nonlinear interactions of electromagnetic forces on charged particles and can therefore be difficult to predict. The perturbations that space storms and other activity causes in some observation sets, however, are fairly regular in their pattern. Some of these measurements have been compiled together into a single value, a geomagnetic index. Several such indices exist, providing a global estimate of the activity in different parts of geospace. Models have been developed to predict the time series of these indices, and various statistical methods are used to assess their performance at reproducing the original index. Existing studies of geomagnetic indices, however, use different approaches to quantify the performance of the model. This document defines a standardized set of statistical analyses as a baseline set of comparison tools that are recommended to assess geomagnetic index prediction models. It also discusses best practices, limitations, uncertainties, and caveats to consider when conducting a model assessment. Key Points: We review existing practices for assessing geomagnetic index prediction models and recommend a "standard set" of metricsAlong with fit performance metrics that use all data‐model pairs in their formulas, event detection performance metrics are recommendedOther aspects of metrics assessment best practices, limitations, uncertainties, and geomagnetic index caveats are also discussed [ABSTRACT FROM AUTHOR]

Abstract Possible climatic effects related to geoeffective solar variability have been investigated by means of long-term statistical correlations between stratospheric and tropospheric temperature and solar/geomagnetic indices. Our previous work on solar variability signature in the long records of air temperature in Europe showed that there were significant solar signals at Schwabe (11 years) and Hale (22 years) solar cycles, with peak to trough amplitudes of several degrees, and, respectively, of 0.6–0.8 °C. In the present study we extend the investigation using NCEP/NCAR reanalyzed data for the temperate climate zone of the Northern Hemisphere (35–65ºN), from Earth's surface to stratospheric levels. Features of these signals are discussed on various spatial scales of the Northern Hemisphere and at specific levels in troposphere and stratosphere. The long-term statistical correlation between reanalyzed temperatures and indices describing solar variability (R, aa) is also investigated. Highlights • Well defined lagged solar signals at 11 and 22-year timescales in reanalyzed temperature. • Identifying statistically significant correlation areas between temperature and R/aa. • Regional rather than global dependence of climate evolution on solar variability. [ABSTRACT FROM AUTHOR]

Statistical wave models, describing the distribution of wave amplitudes as a function of location, geomagnetic activity, and other parameters, are needed as the basis to describe the wave‐particle interactions within numerical models of the radiation belts. In this study, we widen the scope of the statistical wave models by investigating which of the solar wind parameters or geomagnetic indices and their time lags have the greatest influence on the amplitudes of lower band chorus (LBC) waves in the inner magnetosphere. The solar wind parameters or geomagnetic indices with the greatest control over the waves were found using the error reduction ratio (ERR) analysis, which plays a key role in system identification modeling techniques. In this application, the LBC magnitudes at different locations are considered as the output data, while the lagged solar wind parameters are the input data. The ERR analysis automatically determines a set of the most influential parameters that explain the variations in the emissions. Both linear and nonlinear applications of the ERR analysis are compared using solar wind inputs and show that the linear ERR analysis can be misleading. The linear results show that the interplanetary magnetic field (IMF) factor has the most influence on at each magnetic local time (MLT) sector. However, the nonlinear ERR analysis shows that the IMF factor coupled with the solar wind velocity has the main contribution to the LBC wave magnitudes. When geomagnetic indices are included as inputs with the solar wind parameters to the nonlinear ERR analysis, the results show that the majority of the variation in emissions may be attributed to the Auroral Electrojet (AE) index. In the dawn sectors between 00 and 12 MLT and 5 < L < 7, the AE index multiplied by the solar wind velocity with zero time lag has the most influence on the amplitudes of LBC. For 5 < L < 7, the parameters with the highest ERR are the AE index multiplied by the solar wind velocity with a 2‐hr time lag at 12–16 MLT, the linear AE index with a 2‐hr time lag at 16–20 MLT, and AE index multiplied by the IMF factor with zero lag at 20–00 MLT. For 4 < L < 5, the parameters with the highest ERR are the AE index multiplied by the solar wind dynamic pressure with zero time lag at 00–04 MLT, the AE index multiplied by the solar wind velocity with zero time lag between 14 and 12 MLT, the AE index multiplied by the solar wind velocity with a 2‐hr time lag at 12–16 MLT, the Dst index with a 6‐hr time lag at 12–16 MLT, and the AE index multiplied by the IMF factor with zero lag at 20–00 MLT. Plain Language Summary: Lower band chorus (LBC) waves are electromagnetic waves found outside the plasmapause near the geomagnetic equator and are known to modify the local electron distribution through wave‐particle interactions leading to the acceleration of the electrons in the radiation belts to relativistic energies. This study aims to identify the solar wind or geomagnetic drivers of the LBC waves using the error reduction ratio analysis. The error reduction ratio analysis is a system science technique that is able to identify linear and nonlinear combinations of input signals that most influence the output signal. Here the inputs are the measurements of the solar wind and geomagnetic indices, and the output is the LBC wave amplitudes at different locations in the inner magnetosphere. The results show that the majority of the variation in emissions may be attributed to the AE index. The study also shows that solar wind parameters also have a role in the LBC waves. Key Points: Solar wind and geomagnetic drivers of lower band chorus (LBC) waves are determined using the error reduction ratio analysisThe AE index coupled with solar wind velocity is the main driver of the LBC waves on the dawn side between 00‐16 MLTThe time lags of the drivers indicate that these LBC emissions are generated at 00‐12 MLT immediately and with a lag of 2 hr at 12‐14 MLT [ABSTRACT FROM AUTHOR]

In this study, we present a method that combines a Long Short‐Term Memory (LSTM) recurrent neural network with a Gaussian process (GP) model to provide up to 6‐hr‐ahead probabilistic forecasts of the Dst geomagnetic index. The proposed approach brings together the sequence modeling capabilities of a recurrent neural network with the error bars and confidence bounds provided by a GP. Our model is trained using the hourly OMNI and Global Positioning System (GPS) databases, both of which are publicly available. We first develop a LSTM network to get a single‐point prediction of Dst. This model yields great accuracy in forecasting the Dst index from 1 to 6 hr ahead, with a correlation coefficient always higher than 0.873 and a root‐mean‐square error lower than 9.86. However, even if global metrics show excellent performance, it remains poor in predicting intense storms (Dst < −250 nT) 6 hr in advance. To improve it and to obtain probabilistic forecasts, we combine the LSTM model obtained with a GP and evaluate the hybrid predictor using the receiver operating characteristic curve and the reliability diagram. We conclude that this hybrid methodology provides improvements in the forecast of geomagnetic storms, from 1 to 6 hr ahead. Key Points: First use of a Long Short‐Term Memory network to provide single‐point prediction of the Dst index, up to 6 hr aheadDevelopment of a method that combines neural network and Gaussian process to obtain a probabilistic forecast from one to 6 hr aheadUse of specific metrics to evaluate probabilistic forecast, like receiver operating characteristic curves and reliability diagram [ABSTRACT FROM AUTHOR]

The study of the extreme weather space events is important for a technological‐dependent society. Extreme value theory could be decisive to characterize those extreme events in order to have the knowledge to make decisions in technological, economic, and social matters, in all fields with possible impacts. In this work, the hourly values of the Dst geomagnetic index have been studied for the period 1957–2014 using the peaks‐over‐threshold technique. The shape parameter obtained from the fit of the generalized Pareto distribution to the extreme values of the |Dst| index leads to a negative value implying an upper bound for this time series. This result is relevant because the estimation of this limit for the extreme values leads to 850 nT as the highest expected value for this geomagnetic index. Thus, from the previous characterization of the Carrington geomagnetic storm and our results, it could be considered the worst‐case scenario. Plain Language Summary: The study of the space environment that surrounds the Earth is relevant due to the negative effects that unfavorable conditions cause to technological systems. The geomagnetic Dst index is used in this work in order to evaluate the severity of the magnetic storms. The use of statistical tools specifically designed to study rare and scarce extreme events is applied to understand the behavior of this geomagnetic index. The results show that there is a limit value for the geomagnetic Dst index, and therefore, previous registered extreme storms could be considered as the "worst case scenario." Key Points: The shape parameter leads to a negative value implying an upper bound for the |Dst| indexThe estimation of this limit for the extreme values leads to 850 nT as the highest expected value for this geomagnetic indexThe results show that previous registered extreme storms could be considered as the "worst‐case scenario" [ABSTRACT FROM AUTHOR]

We explore the evolution of solar activity for next several sunspot cycles beyond 2016 using data for the geomagnetic indices aa/Ap and the solar polar magnetic field intensity for shorter time intervals; the indices are related to the solar wind and do not depend on Earth's climate. We find that the baseline of the geomagnetic indices increases monotonically from 1900 to 1986 and declines afterwards. We speculate that a cycle with a period ∼86 × 2 = 172y may exist in aa/Ap. If one assumes that solar wind will exhibit the same periodicity for the rest of the twenty-first century, one should expect the next uptick of the aa/Ap timeline to occur in the seventies. In the mean time, the indices Ap/aa may continue to undergo three-cycle-quasi-periodicity (TCQP) to a value lower than in early 1900s, due to a steeper slope during the last few solar cycles compared to that of the period before 1900; it may reach the grand minimum level. Solar polar magnetic field intensity is decreasing systematically for the last three cycles (22–24) as are the sunspot numbers at the cycle peak. Livingston and Penn (2009) note a long term weakening of maximum sunspot magnetic field since 1992. North-South (N-S) asymmetry in the polar field strength is most pronounced for the decay phase of cycles 23, 24; it seems to change sign after cycle 21. These trends have great implications for solar physics and future space weather/climate. We are unable to anticipate the degree and future change of sign of the N-S asymmetry of the solar polar field at present time. [ABSTRACT FROM AUTHOR]

Thermosphere Climate Indexes (TCI) represent the 60-day running average of the global infrared cooling power radiated from the thermosphere by nitric oxide and by carbon dioxide. The TCI are accurately expressed as linear combinations of the 60-day running averages of the F10.7, Ap, and Dst indexes, thus providing terrestrial context to the long record of solar and geomagnetic indexes. We examine the percentile distribution in quintiles of the TCI generated using solar and geomagnetic indexes covering five complete solar cycles. We further assign adjectival descriptors (Cold, Cool, Neutral, Warm, or Hot) to these quintiles as the TCI largely indicate the global thermal state of the thermosphere. We suggest that the TCI are valuable new solar-terrestrial indexes due to the information they contain about the global thermosphere and due to their ease of calculation from standard indexes. Specifically, given dynamic range of the TCI associated with NO cooling, and its significant dependence on both solar irradiance and geomagnetic processes, we recommend that it be included henceforth as a new, standard solar-terrestrial Index. The NO TCI data show that the thermosphere was “Warm” only for a brief period of time at the maximum of solar cycle 24 and thus experienced the coolest solar maximum of the past seven solar cycles. As of February, 2018, the thermosphere power is in the lowest quintile of values, to which we assign the level of ‘Cold.’ [ABSTRACT FROM AUTHOR]

Magnetic storms constitute the most remarkable large-scale phenomena of nonlinear magnetospheric dynamics. Studying the dynamical organization of macroscopic variability in terms of geomagnetic activity index data by means of complexity measures provides a promising approach for identifying the underlying processes and associated time scales. Here, we apply a suite of characteristics from recurrence quantification analysis (RQA) and recurrence network analysis (RNA) in order to unveil some key nonlinear features of the hourly Disturbance storm-time (Dst) index during periods with magnetic storms and such of normal variability. Our results demonstrate that recurrence-based measures can serve as excellent tracers for changes in the dynamical complexity along non-stationary records of geomagnetic activity. In particular, trapping time (characterizing the typical length of “laminar phases” in the observed dynamics) and recurrence network transitivity (associated with the number of the system’s effective dynamical degrees of freedom) allow for a very good discrimination between magnetic storm and quiescence phases. In general, some RQA and RNA characteristics distinguish between storm and non-storm times equally well or even better than other previously considered nonlinear characteristics like Hurst exponent or symbolic dynamics based entropy concepts. Our results point to future potentials of recurrence characteristics for unveiling temporal changes in the dynamical complexity of the magnetosphere. [ABSTRACT FROM AUTHOR]

In this work, the relation of high-latitude indices of geomagnetic activity (AE, Kp) with the rate of storm development and a solar wind electric field during the main phase of magnetic storm induced by the CIR and ICME events is investigated. 72 magnetic storms induced by CIR and ICME events have been selected. It is shown that for the CIR and ICME events the increase of average value of the Kp index (Kp aver ) is observed with the growth of rate of storm development. The value of Kp aver index correlates with the magnitude of minimum value of Dst index (|Dst min |) only for the ICME events. The analysis of average values of AE and Kp indices during the main phase of magnetic storm depending on the SW electric field has shown that for the CIR events, unlike the ICME events, the value of AE aver increases with the growth of average value of the electric field (Esw aver ). The value of Kp aver correlates with the Esw aver only for the ICME events. The relation between geomagnetic indices and the maximum value of SW electric field (Esw max ) is weak. However, for the ICME events Kp aver correlates with Esw max . [ABSTRACT FROM AUTHOR]

Geomagnetic storms, as part of the Sun-Earth relations, are continuously monitored with different indices and scales. These indices have some scale thresholds to quantify the severity or risk of geomagnetic disturbances. However, the most usual scale thresholds are arbitrarily chosen. In this work we aim to quantify the range of the thresholds through a new method. These new thresholds are based on statistical distribution fitting. The data used are from a regional real-time high-cadence geomagnetic index, named LDiñ, and its derivative, LCiñ. We considered the negative part of LDiñ, as significant for geomagnetic disturbances; and the absolute value of LCiñ, significant for geomagnetically induced currents. Then we look for the best fit for different statistical continuous distributions applied to these indices. The method yields that the beta prime is the most suitable functions for negative values of LDiñ, whereas power-law and Johnson-SU are the best fits for LCiñ and the whole distribution, respectively. We define new thresholds for intense, very intense and extreme geomagnetic disturbances as the intersects between these best fit distributions and the index complementary cumulative distribution function. Then, thresholds for the negative part of LDiñ, are -100 nT, -205 and -475 nT. The thresholds for the absolute value of LCiñ, are 6, 18 and 32 nT min-1. The thresholds defined here provide criteria to assess the vulnerability to geomagnetic activity on design or mitigation purposes. These threshold definitions will be applied for different products in the Spanish Space Weather Service (SeNMEs) website. [ABSTRACT FROM AUTHOR]

In this paper, we investigated the interplanetary conditions during 124 less strict high-intensity, long-duration, continuous AE activity (HILDCAA ∗ ) events between the years of 1998 and 2007. The HILDCAA ∗ events were chosen by following three “traditional” criteria of high-intensity, long-duration, continuous AE activity (HILDCAA) events which are characterized with peak of AE intensities equal or greater than 1000 nT; and a minimum of 2 days length where AE values occur outside the main phase of geomagnetic storms. However, we include a small modification in the following criterion: “the AE values should not drop below 200 nT for more than 2 h at a time”. This criterion is modified by changing “2” to “4 h at a time” in which the AE values should not drop below 200 nT. Our results shows that the temporal distribution of HILDCAA ∗ events during the solar cycle presents a pattern of double peak, where the first peak is seen around the rising phase and the maximum of the sunspot cycle 23, with the second peak in its descending phase. This kind of temporal behavior is also observed in HILDCAAs in earlier studies. After the definition of HILDCAA ∗ events, a comparison of solar wind parameters and geomagnetic indices among HILDCAAs, HILDCAAs ∗ , and the background condition is performed using a statistical approach. It is shown that interplanetary causes of HILDCAAs and HILDCAA ∗ s are the same. The advantage of the usage of HILDCAA ∗ s is that the number of events available for study will be ∼3 times higher. [ABSTRACT FROM AUTHOR]

The ionospheric electron density peak ( Nm F2) is analyzed for the recent minima of solar activity for two mid-latitude stations, Rome (41.8°N, 12.5°E, geomagnetic latitude 41.7°N, Italy) and Gibilmanna (37.9°N, 14.0°E, geomagnetic latitude 37.6°N, Italy), and for the low-latitude station of Tucumán (26.9°S, 294.6°E, geomagnetic latitude 17.2°S, Argentina), located in the south ridge of the equatorial ionization anomaly. An inter-minima comparison reveals that from an ionospheric point of view the last minimum of solar activity (minimum 23/24) was peculiar, with values of Nm F2 lower than those recorded during the previous minima for all the stations and all the hours of the day. A more pronounced decrease is observed at Tucumán than at Rome and Gibilmanna. The study of the winter and semi-annual anomaly shows that at mid-latitude stations the winter anomaly is not visible only for the years 2008 and 2009, which represent the deeper part of the prolonged and anomalous last solar minimum. The same is for the semi-annual anomaly. A comparison with the version 2012 of the International Reference Ionosphere model (IRI) is also carried out. The results reveal that for low solar activity the model works better at mid latitudes than at low latitudes, confirming the problems of IRI in correctly representing the low-latitude ionosphere. Nevertheless, using as input updated values of the solar and geomagnetic indices, no loss of accuracy is detected in the IRI performances for the last solar minimum with respect to the previous ones, both at mid and low latitudes. [ABSTRACT FROM AUTHOR]

Measurements of multi satellites and geomagnetic indices are used to investigate the energy budget during an isolated substorm. The calculation of the energy transfer from the solar wind to the magnetosphere (parameter $\varepsilon$ ), the energy increase of the ring current ( $U_{{R}}$ ), the Joule heating ( $U_{{J}}$ ), the particle precipitation energy flux ( $U_{{A}}$ ) and their time-integrated energy dissipation ${W}_{\varepsilon}$ , ${W}_{{R}}$ , ${W}_{{J}}$ , ${W}_{{A}}$ indicates that there should be energy dissipation such as plasma heating and the energy returned to the solar wind by plasmoid ejection from the tail. After calculating the spatial sizes of nine selected BBFs, the energy flux density and energy transported Earthward or tailward by BBFs, using observations from three satellites, are found to be different during an isolated substorm. The flow thermal energy is dominant whether the energy is transported Earthward or tailward under the frozen-in condition in the inner plasma sheet. From results simultaneously observed by three satellites in the magnetotail, we find that the Earthward energy transported by the flows can provide the energy dissipation of ${W}_{{J}}$ and ${W}_{{A}}$ , where the flows are Earthward for more than 60% of the samples, while the tailward energy transport is far larger than ${U}_{{A}}$ and close to ${U}_{{J}}$ , where the flows are tailward for less than 40% of the samples. The maximum energy flux density is observed by one satellite to be accompanied by large variations, while the maximum energy transport is observed by another satellite with large energy flux density and small variations. This suggests misleading conclusions would be obtained if there were only data from single (or two) satellites. From our results, BBFs play an important role in the process of energy transport both Earthward and tailward during this isolated substorm. Data based on observations from one satellite in the magnetotail could be easily misinterpreted and should be used cautiously. [ABSTRACT FROM AUTHOR]

We present a review on geomagnetic indices describing global geomagnetic storm activity ( Kp, am, Dst and dDst/dt) and on indices designed to characterize high latitude currents and substorms ( PC and AE-indices and their variants). The focus in our discussion is in main field modelling, where indices are primarily used in data selection criteria for weak magnetic activity. The publicly available extensive data bases of index values are used to derive joint conditional Probability Distribution Functions (PDFs) for different pairs of indices in order to investigate their mutual consistency in describing quiet conditions. This exercise reveals that Dst and its time derivative yield a similar picture as Kp on quiet conditions as determined with the conditions typically used in internal field modelling. Magnetic quiescence at high latitudes is typically searched with the help of Merging Electric Field ( MEF) as derived from solar wind observations. We use in our PDF analysis the PC-index as a proxy for MEF and estimate the magnetic activity level at auroral latitudes with the AL-index. With these boundary conditions we conclude that the quiet time conditions that are typically used in main field modelling ( $\mathit{PC}<0.8$ , $\mathit{Kp}<2$ and $|\mathit{Dst}|<30~\mbox{nT}$ ) correspond to weak auroral electrojet activity quite well: Standard size substorms are unlikely to happen, but other types of activations (e.g. pseudo breakups $\mathit{AL}>-300~\mbox{nT}$ ) can take place, when these criteria prevail. Although AE-indices have been designed to probe electrojet activity only in average conditions and thus their performance is not optimal during weak activity, we note that careful data selection with advanced AE-variants may appear to be the most practical way to lower the elevated RMS-values which still exist in the residuals between modeled and observed values at high latitudes. Recent initiatives to upgrade the AE-indices, either with a better coverage of observing stations and improved baseline corrections (the SuperMAG concept) or with higher accuracy in pinpointing substorm activity (the Midlatitude Positive Bay-index) will most likely be helpful in these efforts. [ABSTRACT FROM AUTHOR]

The forecasting ability of Multivariate Relevance Vector Machines (MVRVM), used previously to generate forecasts for the Dst index, is extended to forecast the Dst, AL, and PC indices during the years 1975–2007. Such learning machines are used in forecasting because of their robustness, efficiency, and sparseness. The MVRVM model was trained on solar wind and geomagnetic activity data sampled every hour with activity periods of various intensities, durations, and features. It was found that during the training phase, for a given error threshold, 14.60% of the training data was needed to explain the features of the data. The trained model was then tested on 177 different storm intervals, at various levels of geomagnetic activity, to generate simultaneous forecasts of the three indices at a lead time of one hour (1-h). The focus of the modeling was to assess the forecasts during main storm (MS) time periods when the indices show enhanced activity above quiet time values. The forecasts obtained by the MVRVM model reported in this paper returned a MS time average prediction efficiency, PE ¯ of 82.42%, 84.40%, and 76.00% and RMSE ¯ of 13.70 nT, 97.00 nT, and −0.77 mV/m, for the Dst , AL , and PC indices, respectively. The qualitative numbers indicated that the model underestimated the peak amplitude of the indices during the geomagnetic activity, but the peaks were forecasted on time by the model, on average. The forecasting results indicate a robust model generalization and the MVRVM's ability to learn the input-output relationship through a sparse model framework. A qualitative comparison with the previous univariate RVM forecast of Dst indicates that the model goodness of fit numbers improved in the present study. [ABSTRACT FROM AUTHOR]

Recently, we have shown that the contents of total nucleated cells (TNCs) and CD34+hematopoietic stem and progenitor cells (CD34+HSPCs) as well as the cord blood volume (CBV) in umbilical cord blood (UCB) show a circadecadal (~10 years) rhythm of oscillation. This observation was based on an analysis of 17,936 cord blood donations collected during 1999–2011. The aim of the present study was to investigate whether this circadecadal rhythm of oscillation in TNCs, CD34+HSPCs and CBV is related to geomagnetic activity. For the analysis, the yearly averages of TNCs, CD34+HSPCs and CBV in UCB were correlated with geomagnetic activity (Dcxindex). Our analysis revealed that (i) all three UCB parameters were statistically significantly correlated with the level of geomagnetic activity, (ii) CBV showed a linear correlation with theDcxindex (r= 0.5290), (iii) the number of TNCs and CD34+HSPCs were quadratic inversely correlated with theDcxindex (r= −0.5343 andr= −0.7749, respectively). Furthermore, (iv) CBV and the number of TNCs were not statistically significantly correlated with the number of either modest or intense geomagnetic storms per year, but (v) the number of CD34+HSPCs was statistically significantly correlated with the number of modest (r= 0.9253) as well as intense (r= 0.8683) geomagnetic storms per year. In conclusion, our study suggests that UCB parameters correlate with the state of the geomagnetic field (GMF) modulated by solar activity. Possible biophysical mechanisms underlying this observation, as well as the outcome of these findings, are discussed. [ABSTRACT FROM PUBLISHER]

The interplanetary and geomagnetic characteristics of High-Intensity Long-Duration Continuous AE Activity (HILDCAA) events are studied using wavelet analysis technique. The Morlet wavelet transform was applied to the 1 min interplanetary magnetic field (IMF) Bz component and the geomagnetic AE index during HILDCAA events. We have analyzed the AE data for the events occurring between 1975 and 2011, and the IMF Bz data (both in GSE and GSM) for the events between 1995 and 2011. We analyzed the scalograms and the global wavelet spectrum of the parameters. For 50% of all HILDCAA events, the main periodicities of the AE index are generally between 4 and 12 h. For the Bz component, the main periodicities were found to be less than 8 h for ~56% of times in GSM system and for ~54% of times in GSE system. It is conjectured that the periodicities might be associated with the Alfvén waves which have typical periods between 1 and 10 h. The results are discussed in the light of self organized criticality theory where the physical events have the capacity of releasing a considerable amount of energy in a short interval of time. [ABSTRACT FROM AUTHOR]

It is generally accepted that the geomagnetic K indices derived by experienced observers are of great value. The interactive method (IM) based on the traditional hand-scaling methodology is tested in this study. The tests are performed utilising the data from the Hurbanovo and Budkov magnetic observatories. These data include both digital records of the geomagnetic field and hand-scaled K indices that had been derived by experienced observers. The authentic K indices from Hurbanovo cover the year 1997 and the same kind of data from Budkov covers the years 1994–1999. In addition to these data, hand-scaled K indices are used which were derived by the experienced observer from printed digital magnetograms for both of the observatories for the years 2000–2003. The results of this study indicate that for high values of K indices (the values being at least 5) the tested method follows the traditional hand-scaling better than the widely used computer methods FMI and AS. On the other hand, for the K indices less than 5 the tested method turns out to be the worst when compared with the FMI and AS methods. For very low geomagnetic activity ( K -index values equal to 0) the performance of the tested method is comparable to the two computer methods. [ABSTRACT FROM AUTHOR]

On the basis of a morphological analysis of yearly values of the maximum coronal mass ejection (CME) speed index, the sunspot number and total sunspot area, sunspot magnetic field, and solar flare index, the solar wind speed and interplanetary magnetic field strength, and the geomagnetic $A_{\mathrm{p}}$ and $D_{\mathrm{st}}$ indices, we point out the particularities of solar and geomagnetic activity during the last Cycle 23, the long minimum that followed it, and the ascending branch of Cycle 24. We also analyze the temporal offset between the maximum CME speed index and the above-mentioned solar, geomagnetic, and interplanetary indices. It is found that this solar activity index, analyzed jointly with other solar activity, interplanetary parameters, and geomagnetic activity indices, shows a hysteresis phenomenon. It is observed that these parameters follow different paths for the ascending and descending phases of Cycle 23. The hysteresis phenomenon represents a clue in the search for physical processes responsible for linking the solar activity to near-Earth and geomagnetic responses. [ABSTRACT FROM AUTHOR]

This paper proposes a new method to evaluate geomagnetic activity based on wavelet analysis during the solar minimum activity (2007). In order to accomplish this task, a newly developed algorithm called effectiveness wavelet coefficient (EWC) was applied. Furthermore, a comparison between the 5 geomagnetically quiet days determined by the Kp-based method and by wavelet-based method was performed. This paper provides a new insight since the geomagnetic activity indexes are mostly designed to quantify the extent of disturbance rather than the quietness. The results suggest that the EWC can be used as an alternative tool to accurately detect quiet days, and consequently, it can also be used as an alternative to determine the Sq baseline to the current Kp-based 5 quietest days method. Another important aspect of this paper is that most of the quietest local wavelet candidate days occurred in an interval 2 days prior to the high-speed-stream-driven storm events. In other words, the EWC algorithm may potentially be used to detect the quietest magnetic activity that tends to occur just before the arrival of high-speed-stream-driven storms. [ABSTRACT FROM AUTHOR]